Retrieval of soil moisture and vegetation characteristics by use of ERS-1 wind scatterometer over arid and semi-arid areas

The aim of this study is to use the information provided by the ERS-1 wind scatterometer (WSC) over land surfaces in arid and semi-arid environments to infer soil moisture in the presence of vegetation. Driven by dielectric properties and surface roughness, the soil contribution is attenuated by a factor which depends on canopy characteristics (water content, shape, height, density) and scatterometer viewing conditions. To describe the influence of vegetation on the signal, a semiempirical 'water-cloud' model (a first-order radiative transfer solution) was used. The optical thickness (tau) and the single scattering albedo (omega) are the parameters used to quantify vegetation contribution to the measured signal. Through a simulation analysis for different soil moisture and viewing (incidence angle) conditions, we show the importance of tau and omega on the signal partition between vegetation and soil contributions. To quantify the effect of vegetation on the signal, we used information on the green vegetation acquired from NOAA-AVHRR, visible and near-IR data combined with WSC satellite data in a water-cloud model to extract tau and omega. The temporal evolution of the various contributions to the signal was then compared for different angular ranges. This semiempirical model was then applied within suitable angular ranges to retrieve soil moisture.